U.S. patent application number 16/146875 was filed with the patent office on 2019-01-31 for methods and apparatus to partition data.
The applicant listed for this patent is The Nielsen Company (US), LLC. Invention is credited to Peter Lipa, Michael Sheppard, Jonathan Sullivan, Alejandro Terrazas.
Application Number | 20190034492 16/146875 |
Document ID | / |
Family ID | 58282846 |
Filed Date | 2019-01-31 |
![](/patent/app/20190034492/US20190034492A1-20190131-D00000.png)
![](/patent/app/20190034492/US20190034492A1-20190131-D00001.png)
![](/patent/app/20190034492/US20190034492A1-20190131-D00002.png)
![](/patent/app/20190034492/US20190034492A1-20190131-D00003.png)
![](/patent/app/20190034492/US20190034492A1-20190131-D00004.png)
![](/patent/app/20190034492/US20190034492A1-20190131-D00005.png)
![](/patent/app/20190034492/US20190034492A1-20190131-D00006.png)
United States Patent
Application |
20190034492 |
Kind Code |
A1 |
Sheppard; Michael ; et
al. |
January 31, 2019 |
METHODS AND APPARATUS TO PARTITION DATA
Abstract
Methods and apparatus to partition data are disclosed. An
example apparatus to partition panelist data includes an indicator
matrix generator to determine an indicator matrix including
panelist vectors for panelists based on panelist data associated
with the panelists, a first one of the panelist vectors for a first
one of the panelists to indicate whether the first one of the
panelists: has a first characteristic, meets a first criterion
based on the first characteristic and a second characteristic, has
a third characteristic, and meets a second criterion based on the
third characteristic and a fourth characteristic. The example
apparatus further includes a matrix reducer to reduce the indicator
matrix to a set of unique panelist vectors that represent
partitions of the panelist data, the partitions of the panelist
data to utilize less storage capacity than the panelist data.
Inventors: |
Sheppard; Michael;
(Brooklyn, NY) ; Sullivan; Jonathan; (Hurricane,
UT) ; Lipa; Peter; (Tucson, AZ) ; Terrazas;
Alejandro; (Santa Cruz, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
The Nielsen Company (US), LLC |
New York |
NY |
US |
|
|
Family ID: |
58282846 |
Appl. No.: |
16/146875 |
Filed: |
September 28, 2018 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
14860361 |
Sep 21, 2015 |
10089358 |
|
|
16146875 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06Q 30/0201 20130101;
G06F 16/24556 20190101; G06F 16/2465 20190101; G06F 16/2365
20190101 |
International
Class: |
G06F 17/30 20060101
G06F017/30; G06Q 30/02 20120101 G06Q030/02 |
Claims
1. An apparatus to partition panelist data, the apparatus
comprising: an indicator matrix generator to determine an indicator
matrix including panelist vectors for panelists based on panelist
data associated with the panelists, a first one of the panelist
vectors for a first one of the panelists to indicate whether the
first one of the panelists: has a first characteristic; meets a
first criterion based on the first characteristic and a second
characteristic; has a third characteristic; and meets a second
criterion based on the third characteristic and a fourth
characteristic; and a matrix reducer to reduce the indicator matrix
to a set of unique panelist vectors that represent partitions of
the panelist data, the partitions of the panelist data to utilize
less storage capacity than the panelist data.
2. The apparatus of claim 1, further including a panelist labeler
to assign labels to the unique panelist vectors.
3. The apparatus of claim 2, wherein the matrix reducer is to:
store the set of unique panelist vectors and the labels assigned to
the unique panelist vectors; and discard the panelist data.
4. The apparatus of claim 2, wherein the panelist labeler is to
attach respective descriptions to the unique panelist vectors, the
descriptions indicating whether respective ones of the unique
panelist vectors have the first characteristic, the second
characteristic, the third characteristic and the fourth
characteristic.
5. The apparatus of claim 2, wherein the panelist labeler is to
assign panelist labels to the panelists, the panelist labels
corresponding to the labels assigned to the unique panelist
vectors.
6. The apparatus of claim 5, further including a panelist sorter to
sort the panelists into groups corresponding to the panelist labels
assigned to the panelists.
7. The apparatus of claim 6, wherein the panelist sorter is to:
determine numbers of panelists included in the respective groups;
and output processed panelist data indicative of numbers of
panelists associated with respective ones of the labels, the
processed panelist data to be used in weighting the partitions to
represent a population.
8. The apparatus of claim 1, wherein the indicator matrix generator
is to determine the first criterion and the second criterion based
on characteristics exhibited by a threshold number of the
panelists.
9. A method to partition panelist data, the method comprising:
determining, by executing an instruction with a processor, an
indicator matrix including panelist vectors for panelists based on
panelist data associated with the panelists, a first one of the
panelist vectors for a first one of the panelists to indicate
whether the first one of the panelists: has a first characteristic;
meets a first criterion based on the first characteristic and a
second characteristic; has a third characteristic; and meets a
second criterion based on the third characteristic and a fourth
characteristic; and reducing, by executing an instruction with the
processor, the indicator matrix to a set of unique panelist vectors
that represent partitions of the panelist data, the partitions of
the panelist data to utilize less storage capacity than the
panelist data.
10. The method of claim 9, further including assigning labels to
the unique panelist vectors.
11. The method of claim 10, further including: storing the set of
unique panelist vectors and the labels assigned to the unique
panelist vectors; and discarding the panelist data.
12. The method of claim 10, further including attaching respective
descriptions to the unique panelist vectors, the descriptions
indicating whether respective ones of the unique panelist vectors
have the first characteristic, the second characteristic, the third
characteristic and the fourth characteristic.
13. The method of claim 10, further including assigning panelist
labels to the panelists, the panelist labels corresponding to the
labels assigned to the unique panelist vectors.
14. The method of claim 13, further including sorting the panelists
into groups corresponding to the panelist labels assigned to the
panelists.
15. The method of claim 9, further including determining the first
criterion and the second criterion based on characteristics
exhibited by a threshold number of the panelists.
16. A tangible computer readable storage medium comprising
instructions that, when executed, cause a machine to at least:
determine an indicator matrix including panelist vectors for
panelists based on panelist data associated with the panelists, a
first one of the panelist vectors for a first one of the panelists
to indicate whether the first one of the panelists: has a first
characteristic; meets a first criterion based on the first
characteristic and a second characteristic; has a third
characteristic; and meets a second criterion based on the third
characteristic and a fourth characteristic; and reduce the
indicator matrix to a set of unique panelist vectors that represent
partitions of the panelist data, the partitions of the panelist
data to utilize less storage capacity than the panelist data.
17. The tangible computer readable storage medium of claim 16,
wherein the instructions further cause the machine to assign labels
to the unique panelist vectors.
18. The tangible computer readable storage medium of claim 17,
wherein the instructions further cause the machine to: store the
set of unique panelist vectors and the labels assigned to the
unique panelist vectors; and discard the panelist data.
19. The tangible computer readable storage medium of claim 17,
wherein the instructions further cause the machine to attach
respective descriptions to the unique panelist vectors, the
descriptions indicating whether respective ones of the unique
panelist vectors have the first characteristic, the second
characteristic, the third characteristic and the fourth
characteristic.
20. The tangible computer readable storage medium of claim 17,
wherein the instructions further cause the machine to assign
panelist labels to the panelists, the panelist labels corresponding
to the labels assigned to the unique panelist vectors.
Description
RELATED APPLICATION
[0001] This patent arises from a continuation of U.S. patent
application Ser. No. 14/860,361, (Now U.S. Pat. No. ______) which
was filed on Sep. 21, 2015. U.S. patent application Ser. No.
14/860,361 is hereby incorporated herein by reference in its
entirety. Priority to U.S. patent application Ser. No. 14/860,361
is hereby claimed.
FIELD OF THE DISCLOSURE
[0002] This disclosure relates generally to data segmentation and,
more particularly, to methods and apparatus to partition panelist
data.
BACKGROUND
[0003] Market research groups collect data to evaluate the market
for certain products and determine how the demographics of a
population may affect the market. To collect such research data,
market researchers may employ a group of selected participants
(known as panelists) to provide data, such as demographic data,
media exposure data, and/or other data related to the market
research. Each panelist may be selected such that the plurality of
panelists is an accurate scale representation of one or more larger
populations that a market researcher is studying. Because the
research data from the panelists is based on a relatively small
group of people, inaccuracies sometimes exist in scaling the data
from the panelists to the larger population of interest. To
decrease the effect of these inaccuracies as the market research is
applied to the larger population(s), the research data may be
processed, for example, by weighing certain groups of panelists
differently than other groups of panelists.
BRIEF DESCRIPTION OF THE DRAWINGS
[0004] FIG. 1 is a block diagram of an example environment in which
an example data partitioner constructed in accordance with the
teachings of this disclosure operates to partition panelist data
based on input criteria.
[0005] FIG. 2 is a block diagram of an example implementation of
the indicator matrix generator of FIG. 1.
[0006] FIG. 3 is a flowchart representative of example machine
readable instructions that may be executed to implement the example
data partitioner of FIG. 1.
[0007] FIG. 4 is a flowchart representative of example machine
readable instructions that may be executed to implement the example
indicator matrix generator of FIGS. 1 and/or 2.
[0008] FIG. 5 is an example set of matrix criteria that may be
received by the example data partitioner of FIG. 1.
[0009] FIG. 6 is an example indicator matrix that may be generated
by the example indicator matrix generator of FIGS. 1 and/or 2.
[0010] FIG. 7 is an example set of partitioned data that may be
determined by the example data partitioner of FIG. 1.
[0011] FIG. 8 is an example set of panelist data that is matched in
accordance with the partitions of FIG. 7.
[0012] FIG. 9 is an example set of panelist data that is sorted in
accordance with the partitions of FIG. 7.
[0013] FIG. 10 illustrates an example set of partition
descriptions.
[0014] FIG. 11 is a block diagram of an example processor platform
structured to execute the example machine-readable instructions of
FIGS. 3 and 4 to implement the example data partitioner and/or
indicator matrix generator of FIGS. 1 and/or 2.
[0015] Wherever possible, the same reference numbers will be used
throughout the drawing(s) and accompanying written description to
refer to the same or like parts.
DETAILED DESCRIPTION
[0016] Example methods and apparatus disclosed herein construct a
minimum set of mutually exclusive partitions of panelist data based
on an identification of matrix criteria to be studied. In some
examples, responses to all matrix criteria being studied are
represented by (e.g., can be determined from) the minimum set of
mutually exclusive partitions. The matrix criteria may be
represented by probabilities that evaluate the panelist for one
characteristic and/or conditional probabilities that evaluate the
panelists for one characteristic given another characteristic. For
example, a matrix criterion may be the conditional probability that
a panelist is male given that the panelist has a cell phone (e.g.,
represented as P(Male Cell Phone)), where "male" is a first
characteristic (e.g., an event) and "cell phone" is a second
characteristic (e.g., a condition).
[0017] As used herein, the term "event" refers to the tested
portion of the probability (e.g., Male is the event in the
following probabilities: P(Male) and P(Male|Cell Phone). As used
herein, the term event is not limited to happenings or occurrences,
but rather includes anything (e.g., occurrence, trait, attribute,
etc.) that may be tested by a probability. For example, an event as
used herein may include a characteristic of a panelist such as an
age, gender, income, etc. or may be a behavior (e.g., purchase
behavior, media access behavior, etc.) As used herein, the term
condition refers to the second half of a condition probability
(e.g., the portion to the right of the "|"). In the following
disclosure, the term condition may also be referred to as the
"given" or the "given characteristic."
[0018] As used herein, a minimum set of mutually exclusive
partitions is the set including the least number of groupings of
panelist attributes that may be used to calculate the probabilities
(e.g., including conditional probabilities) included in received
matrix criteria. For example, if data regarding two characteristics
is collected (e.g., gender of the panelist and whether or not the
panelist has a cell phone), the mutually exclusive partitions
include all possible outcomes present in the sample population
(e.g., male with cell phone, male without cell phone, female with
cell phone, and female without cell phone). However, if, for
example, the only matrix criterion to be studied is
P(age>20|Cell Phone), the matrix criterion can be determined
with knowledge of the number of panelists older than 20 with cell
phones and the total number of persons (e.g., the sum of the number
of all panelists). Thus, in such an example, the minimum set of
mutually exclusive partitions would be a partition including
panelists older than 20 with a cell phone and a partition including
panelists 20 or younger with a cell phone and panelists without a
cell phone.
[0019] In some examples, after determining the minimum set of
mutually exclusive partitions, the full set of data for the
panelists may be discarded in favor of retaining the data for the
minimum set of mutually exclusive partitions (e.g., a data set
having 10,000 elements (e.g., 1000 panelist partitions each having
10 data elements for a total of 10,000 elements) may be reduced to,
for example, four elements that may be used to analyze two matrix
criteria).
[0020] An example apparatus to partition data as disclosed herein
may receive data corresponding to a plurality of panelists and
store the data in a panelist data storage device. Some such example
apparatus also include an indicator matrix generator to generate an
indicator matrix from which the mutually exclusive partitions are
determined. In some examples, the example indicator matrix
generator includes a parameter definer to define the matrix
parameters of the indicator matrix based on the panelists and the
matrix criteria received by the example apparatus. The example
indicator matrix generator of some examples also includes a
criteria analyzer to determine if the panelists meet the matrix
criteria and an element and vector positioner to place an indicator
in a vector of the indicator matrix corresponding to a matrix
criterion if the panelist meets the matrix criterion. Any
indicators placed in a matrix criterion vector by the element and
vector positioner will correspond to the panelist being analyzed by
the criteria analyzer.
[0021] The example apparatus of some examples also includes a
matrix reducer to reduce the indicator matrix to the minimum set of
mutually exclusive partitions. Some such example apparatus also
include a panelist labeler to label the partitions (e.g., A, B, C,
D, etc.). In some examples, the apparatus includes a panelist
sorter to sort the panelists into mutually exclusive partitions
corresponding to the matrix criteria met by each panelist. In some
examples, the number of panelists sorted into each mutually
exclusive partition is determined.
[0022] Example methods disclosed herein include receiving panelist
data from a panelist data storage device, generating an indicator
matrix, determining mutually exclusive partitions from the
generated indicator matrix, assigning each mutually exclusive
partition a label, matching each panelist to a corresponding
mutually exclusive partition, assigning labels to the panelists
corresponding to the mutually exclusive partition to which each
panelist was matched, and sorting the panelists according to the
assigned label. Some such example methods define the indicator
matrix parameters, determine if a given panelist meets a given
matrix criterion, and place an indicator in the corresponding
position in the indicator matrix if the panelist meets the given
matrix criterion. In some example methods, generating the indicator
matrix is repeated for each combination of panelist and matrix
criterion to populate the indicator matrix.
[0023] FIG. 1 is a block diagram of an example environment 100 in
which an example data partitioner 106 may be utilized to partition
panelist data. The example environment 100 includes a plurality of
example panelists 102 from which panelist data is collected. The
collected data is stored in an example panelist data storage device
104. The example environment 100 also includes an example data
partitioner 106 to analyze the panelist data in the example
panelist data storage device 104 and produce processed panelist
data 108.
[0024] As used herein, panelist data includes any information
collected about the example panelists 102 (e.g., demographics,
behaviors, etc.). In some examples illustrated herein, the
information is represented as Boolean, categorical, or ordinal data
sets. In some examples including data sets with real numbers,
brackets are created (e.g., an age bracket of 18-24, an age bracket
25-35, etc.). For example, the panelist data may include
characteristics of the panelist such as, for example, demographic
information, product preferences, television viewing preferences,
product ownership information, etc. For example, demographic
information may be collected at the time of panelist registration.
In some examples, demographic information includes gender, age,
income, location, presence of children, ownership of a device, etc.
Other information included in the panelist data, such as product
preferences and television viewing preferences, may be collected
during panelist registration by requesting that the panelists
complete a survey and/or by automated media measurement device(s)
associated with the panelist(s) 102, such as the portable
meter.
[0025] The example plurality of panelists 102 may include any
number of panelists representative of a target population. In the
illustrated example, panelist data is collected from the example
panelists 102 using portable meters that have been given to the
panelists 102 for the purpose of data collection. In some examples,
each panelist 102 is respectively assigned a portable meter. In
some examples, the portable meters may be integrated into a mobile
device (e.g., a cellular phone) carried by the panelists 102. In
other examples, the portable meter may be a wearable device, such
as a watch. Alternatively, the panelist data may be collected via
surveys sent to the panelists 102. In some such examples, the
surveys may be sent to the panelists 102 via the Internet (e.g.,
email), physical mailing, etc. In other examples, the panelist data
may be collected through other measurement devices and/or methods,
such as by collection of data from set top boxes utilized by the
example panelists 102.
[0026] The panelists 102 of the example of FIG. 1 are users
registered on panels maintained by a ratings entity (e.g., an
audience measurement company) that owns and/or operates the ratings
entity subsystem. Traditionally, audience measurement entities
(also referred to herein as "ratings entities") determine
demographic reach for advertising and/or media programming based on
registered panel members. That is, an audience measurement entity
enrolls people that consent to being monitored into a panel. During
enrollment, the audience measurement entity receives demographic
information from the enrolling panelists so that determinations can
be made about the composition of the audience experiencing the
advertisement/media exposure, the effectiveness of the
advertisements, and/or the behavior of the audience members.
[0027] Panelists 102 may be enrolled in the panel via, for example,
a user interface presented on a media device (e.g., via a website).
Panelists 102 may be enrolled in the panel in additional or
alternative manners such as, for example, via a telephone
interview, by completing an online survey, random selection,
statistical selection, phone solicitations, Internet
advertisements, surveys, advertisements in shopping malls, product
packaging, etc.
[0028] The example panelist data storage device 104 of the example
of FIG. 1 is a server operative to collect data (including panelist
data) from the example portable meters carried by the example
panelists 102. The example panelist data storage device 104 of FIG.
1 is in communication with the example panelists 102 (e.g., with
portable meters carried by the example panelists 102) via either a
wired connection or via a wireless connection, such as wireless
Internet, Bluetooth, Near Field Communication (NFC), or RFID (Radio
Frequency Identification), etc. Alternatively, the panelist data
storage device 104 may be implemented by a database, a hard drive,
hard disk, or any other type of tangible computer readable medium,
such as a compact disk or flash drive. The example panelist data
storage device 104 may be in communication with the example data
partitioner 106 to facilitate analysis of collected panelist data
by the example data partitioner 106.
[0029] As used herein, the phrase "in communication," including
variances thereof, encompasses direct communication and/or indirect
communication through one or more intermediary components and does
not require direct physical (e.g., wired) communication and/or
constant communication, but rather additionally includes selective
communication at periodic or aperiodic intervals, as well as
one-time events.
[0030] The data partitioner 106 of the illustrated example of FIG.
1 analyzes panelist data collected from the example panelists 102
to partition the panelist data into mutually exclusive partitions.
In the illustrated example, matrix criteria are provided to the
example data partitioner 106 as input from a user of the example
data partitioner 106 (e.g., an administrator of a study). In other
examples, the data partitioner 106 may determine the matrix
criteria based on the given set of panelist data by, for example,
determining characteristics that many of the example panelists 102
have in common.
[0031] As used herein, the term matrix criteria refers to a set of
one or more matrix criterion (e.g., a single matrix criterion,
multiple matrix criterion, etc.) that is used to evaluate a set of
panelist data. As used herein, a matrix criterion is a probability
or a conditional probability that a panelist has one or more
characteristics. As used herein, a characteristic is an attribute
of the panelist data to be evaluated. For example, a probability
may include two characteristics (e.g., P(first
characteristic|second characteristic). In some such examples, the
first characteristic may be referred to as an event and the second
characteristic may be referred to as a condition (e.g.,
P(event|condition). The terms first characteristic and event may be
used interchangeably and likewise, the terms second characteristic
and condition may be used interchangeably. As used herein, an
attribute of a panelist is any information about a panelist (e.g.,
demographics, behavior, etc.) or otherwise associated with a
panelist. For example, an attribute may be demographic information
about a panelist, an answer that the panelist provided in response
to a survey, a purchase behavior, an activity, a preference, etc.
In some examples, the event and/or the condition may include one or
more attributes to be evaluated (e.g., gender and age).
[0032] In the illustrated example, the matrix criteria provided to
the data positioner are represented as probabilities including only
one characteristic (e.g., P(Male). In such examples, the
characteristic is referred to as the event. In other examples, the
matrix criteria provided to the data partitioner 106 are
represented as conditional probabilities (e.g., P(Male|Cell Phone))
and the matrix criteria includes two characteristics (e.g., male,
cell phone) the user wishes to evaluate for the panelist data. For
example, the first characteristic (e.g., male) may be the event of
the probability and the second characteristic (e.g., whether the
panelist has a cell phone) may be the condition of the probability.
In examples where the characteristic(s) includes more than one
attribute to be evaluated, the matrix criteria may be represented
as, for example, P(Male.andgate.age<20|Cell Phone). In this
example, the event (e.g., male and age is less than 20) includes
two attributes of the panelist. Other examples where the
characteristic(s) include more than one attribute may include
P(Male.andgate.Cell Phone), P(Male|Cell Phone.andgate.age<20),
include P(Male.andgate.location=USA|Cell Phone.andgate.age<20),
P(Male.orgate.age<20|Cell Phone), etc.
[0033] To partition the panelist data stored in the example
panelist data storage 104, the example data partitioner 106 creates
an indicator matrix based on the matrix criteria received by the
example data partitioner 106. The indicator matrix includes a
panelist vector for each panelist of the plurality of panelists
102. In the illustrated example, the panelist vector includes an
indicator for each of the event and the condition of each matrix
criterion in the set of matrix criteria. If a matrix criterion
includes only an event (e.g., a probability such as P(Male) that is
not a conditional probability), the panelist vector includes a
first indicator for the event and a second indicator for the
condition, which evaluates as true for all panelists in the set of
panelists. The example data partitioner 106 of FIG. 1 reduces the
indicator matrix to a set of unique vectors, which corresponds to a
minimum set of mutually exclusive partitions. The example data
partitioner 106 of FIG. 1 also assigns a label (e.g., A, B, C, or
D) to each unique vector and assigns a corresponding label to each
panelist vector. The example data partitioner 106 sorts the example
panelists 102 based on the corresponding label assigned to the
panelist vector corresponding to each of the panelists 102.
[0034] In further detail, the example data partitioner 106 of FIG.
1 includes an example indicator matrix generator 110, an example
matrix reducer 112, an example panelist labeler 114 and an example
panelist sorter 116.
[0035] The example indicator matrix generator 110 of FIG. 1
generates the indicator matrix based on collected data for the
plurality of panelists 102 and the given matrix criteria. An
example indicator matrix 604 is shown and described in more detail
in conjunction with FIG. 6. The indicator matrix generator 110 of
the example defines the matrix parameters of the indicator matrix,
evaluates each of the panelists 102 based on the provided matrix
criteria and the corresponding panelist data, and places an
indicator, such as indictor 636 or indicator 638 of FIG. 6, in the
corresponding location of the panelist vector in the indicator
matrix. An example panelist vector 640 is shown in the example
indicator matrix 604 of FIG. 6. As used herein, a matrix parameter
is the number of rows and/or columns in the indicator matrix (e.g.,
a two dimensional indicator matrix may have two matrix parameters
(a first matrix parameter identifying the number of rows and a
second matrix parameter identifying the number of columns)). Thus,
defining the matrix parameters of the indicator matrix includes
determining the number of rows and columns needed to input each of
the panelists for the given set of matrix criteria in the indicator
matrix. The example panelist vectors include an indication of
whether a corresponding panelist fulfils each of the given matrix
criteria. If two characteristics exist in a matrix criterion, the
indicator matrix generator 110 evaluates the matrix criterion by
determining the probability of whether or not each of the panelists
102 fulfills the event, given the condition (e.g., out of the
panelists who fulfill the condition, which also fulfill the event).
Thus, the example indicator matrix generator 110 of FIG. 1
generates an indicator matrix where the panelist vector
corresponding to each of the panelists 102 includes two indicators
for each evaluated matrix criterion.
[0036] In the illustrated example, each of the matrix criteria
evaluated by the example indicator matrix generator 110 and
represented in the indicator matrix is assigned two indicators per
panelist. The first indicator corresponds to whether the panelist
fulfills the condition (e.g., the second characteristic) and the
second indicator corresponds to whether the panelist fulfills both
the event and the condition (e.g., has both the characteristics).
If only one characteristic (e.g., the event) is defined by a matrix
criterion, the example indicator matrix generator 110 evaluates the
matrix criterion by determining whether or not each of the
panelists 102 has that characteristic. In such cases where there is
only one characteristic in a matrix criterion, the condition is
whether the panelist is a part of the sample (e.g., if the panelist
is in the universe), which is a characteristic of every panelist
being evaluated and thus every panelist would include an indicator
that indicates that they are included in the set.
[0037] The evaluation of the matrix criteria by the example
indicator matrix generator 110 includes two possible outcomes for
each component (e.g., the event and the condition) of each matrix
criterion: (1) yes, the panelist has the characteristic(s) and
meets the component of the matrix criterion or (2) no, the panelist
does not meet the component of the matrix criterion. An answer of
"yes" may correspond to a first type of indicator being placed in
the matrix, for example, a "1," and an answer of "no" may
correspond to a second type of indicator being placed in the
matrix, for example, a "0." The example indicator matrix generator
110 populates the indicator matrix with the different types of
indicators for each of the panelists 102 corresponding to the
result of the evaluation of the matrix criteria. Each of the
panelists 102 corresponds to a panelist vector (e.g., a row) which
contains a position (e.g., a column) in the vector for the
determined indicators for each of the matrix criteria. Each column
defines an element vector such that the indicators in the element
vector correspond to a same matrix criterion. The indicator matrix
created by the example indicator matrix generator 110 is a
compilation of all of the panelist vectors for each of the
panelists 102 in the sample set. The example indicator matrix
generator 110 is described in further detail in conjunction with
FIG. 2 below.
[0038] The example matrix reducer 112 of the data partitioner 106
of FIG. 1 reduces the indicator matrix generated by the example
indicator matrix generator 110 to the set of unique panelist
vectors (e.g., unique rows of indicators). For example, the matrix
reducer 112 evaluates the panelist vectors to determine a set of
unique panelist vectors representative of all of the panelist
vectors in the indicator matrix. The set of unique vectors
determined by the matrix reducer 112 corresponds to the minimum set
of mutually exclusive partitions for the given set of panelist
data.
[0039] The minimum set of mutually exclusive partitions may vary
based on the panelists 102 and the given matrix criteria. For
example, two sets of panelist data evaluated using the same matrix
criteria may result in two different minimum sets of mutually
exclusive partitions. In some such examples, if the matrix criteria
being evaluated include gender of the panelist and whether or not
the panelist has a cell phone, the mutually exclusive partitions
may include "male with cell phone," "male without cell phone,"
"female with cell phone," "and female without cell phone," or any
subset of those partitions listed, depending on the panelist data.
In some examples, a data set may not include any panelists that
fulfill a particular combination of the matrix criteria and, thus,
that combination of matrix criteria would not be represented in the
minimum set of mutually exclusive partitions for that data set. For
example, the data may not contain a male panelist without a cell
phone and, thus, the "male without cell phone" partition would not
be included among the set of mutually exclusive partitions. The
minimum set of mutually exclusive partitions can be used to
reproduce the data from the sample of panelists 102. For example,
any information a user wishes to obtain from the panelist data may
be obtained using the set of mutually exclusive partitions because
every outcome of panelist data present in the sample is represented
in the minimum set of mutually exclusive partitions. In some
examples, the matrix reducer 112 may, instead, reduce the indicator
matrix to a set of panelist vectors matching a specific outcome.
For example, the user may only want to see the partitions for
panelists that have a cell phone.
[0040] The example panelist labeler 114 of FIG. 1 receives the
reduced indicator matrix (e.g., the unique vectors representing the
minimum set of mutually exclusive partitions) from the example
matrix reducer 112 and assigns each of the unique vectors in the
minimum set of mutually exclusive partitions a corresponding label.
In the illustrated example, the mutually exclusive partitions are
labeled with a letter (A, B, C, D, etc.). In other examples, the
label may be a number, a word, and/or a combination of words,
numbers, letters, symbols, etc. The matrix labeler 114 of the
illustrated example also labels any panelist vectors matching a
unique vector with a corresponding label. In some examples, the
example panelist labeler 114 sorts the panelist vectors according
to matching unique vectors prior to the panelist vectors being
labeled by the example panelist labeler 114. Additionally, the
matrix labeler 114 may attach a description to each of the
previously labeled mutually exclusive partitions to indicate the
characteristics of the panelists 102 belonging to each mutually
exclusive partition.
[0041] The example panelist sorter 116 of FIG. 1 sorts the labeled
panelist vectors into groups according to the label assigned to
each of the panelist vectors. For example, the matrix sorter 116
may sort the panelists 102 to enable more efficient subsequent
processing of the panelist data by allowing the user to select
which partitions are to be presented as processed panelist data
108. In some examples, the example matrix sorter 116 may sort the
panelists 102 into groups and determine the number of panelists 102
in each group. The panelist sorter 116 outputs a list of processed
panelist data 108 based on the sorted groups of panelists 102. In
some examples, input from the user indicates how the panelist
sorter 116 presents the list of processed panelist data 108. In
other examples, the processed panelist data 108 may include all of
the labeled panelist vectors.
[0042] The processed panelist data 108 of the illustrated example
is a list or table of the minimum set of mutually exclusive
partitions. Each mutually exclusive partition may be paired with a
description indicating the type of panelists 102 that are included
in the partition. In some examples, the processed panelist data 108
may be the entire indicator matrix. Alternatively, the processed
panelist data may include any subset of the indicator matrix, such
as all the panelists 102 in a specific partition. The processed
panelist data 108 may be used in other data processing methods that
may include weighting the partitions to obtain a more accurate
representation of a larger population (e.g., the universe) based on
the sample of panelists 102. Some such example methods, apparatus,
and systems that utilize partitioned data are disclosed in U.S.
patent application Ser. No. 14/835,401, filed Aug. 25, 2015
entitled Reducing Processing Requirements to Correct for Bias in
Ratings Data Having Interdependencies Among Demographic Statistics,
which is hereby incorporated by reference in its entirety.
[0043] In operation the example panelist data storage 104 receives
data from the example plurality of panelists 102. Data from the
example panelists 102 may be gathered using one or more methods
(e.g., portable meter devices, surveys, set top boxes instrumented
with meters, etc.). The data gathered from the example panelists
102 is stored in the example panelist data storage device 104. To
enable the panelist data to be used effectively, the panelist data
is evaluated by the example data partitioner 106. In particular,
the example indicator matrix generator 110 of the data partitioner
106 evaluates the panelist data by generating an indicator matrix.
The example indicator matrix generator 110 defines the matrix
parameters (e.g., the number of rows and columns for the indicator
matrix based on the set of matrix criteria and the number of
panelists 102) of the indicator matrix, determines indicators for
each of the panelists 102 based on each of the matrix criteria, and
populates the indicator matrix with the indicators. The example
matrix reducer 112 reduces the matrix to a minimum set of mutually
exclusive partitions. The mutually exclusive partitions contain all
of the combinations of the matrix criteria under evaluation for the
given population of the example panelists 102. The example panelist
labeler 114 then labels the partitions (e.g., using the labels A,
B, C, D, etc.). In some examples, the panelists 102 in each
mutually exclusive partition are also labeled accordingly (e.g.,
using corresponding labels A, B, C, D, etc.). The example panelist
sorter 116 sorts the panelists 102 based on the label(s) to output
the example processed panelist data 108.
[0044] FIG. 2 is a block diagram of an example implementation of
the indicator matrix generator 110 of FIG. 1. The example indicator
matrix generator 110 shown in FIG. 2 includes an example parameter
definer 202, an example criteria analyzer 204, and an example
element and vector positioner 206.
[0045] The example parameter definer 202 of FIG. 2 defines matrix
parameters of an indicator matrix based on a number of the received
matrix criteria and a number of the example panelists in a
plurality of panelists 102. In the illustrated example, the
parameter definer 202 assigns a row (e.g., a panelist vector) in
the indicator matrix for each of the example panelists 102. The
example parameter definer 202 determines the number of columns
(e.g., element vectors) of the indicator matrix based on the number
of provided matrix criteria, with each matrix criterion being
assigned two columns. The first element vector for a matrix
criterion corresponds to whether each of the panelists 102 has a
first characteristic (e.g., the condition of a probability)
associated with the matrix criterion and a second element vector
for the matrix criterion corresponds to whether each of the
panelists 102 has the first characteristic (e.g., the condition of
a probability) associated with the matrix criterion and the second
characteristic (e.g., the event). In some examples, such as the
example illustrated herein in FIG. 6, the element vector 622 for
matrix criterion 502 may be located to the right of the second
element vector 620 for matrix criterion 502 in an example indicator
matrix 604.
[0046] The example criteria analyzer 204 of FIG. 2 evaluates each
of the panelists 102 based on the received matrix criteria (which
are/is a probability or a conditional probability) to determine
whether or not the panelist has the first characteristic and/or the
second characteristic. In the illustrated example, each matrix
criterion has an associated first indicator corresponding to
whether a panelist has the first characteristic (e.g., the
condition) of a probability and an associated second indicator
corresponding to whether the panelist has the first characteristic
and the second characteristic (e.g., the event and the condition)
of the example matrix criterion. The criteria analyzer 204
determines the first indicator based on whether or not the panelist
has the second characteristic (e.g., the condition). The example
criteria analyzer 204 determines the second indicator based on
whether or not the panelist has the first characteristic and the
second characteristic (e.g., the event and the condition). In other
examples, each of the matrix criteria may only have one
corresponding indicator. In such examples, the criteria analyzer
204 may determine the indicator based on whether the panelist has
both the first and the second characteristics. In such examples, if
the panelist does not have both the first and second
characteristics, the panelist is assigned an indicator indicating
that the panelist does not meet the corresponding matrix
criterion.
[0047] The example element and vector positioner 206 of the
illustrated example places an indicator (e.g., a 1 or a 0) in a
location of the indicator matrix corresponding to the panelist and
the matrix criterion being evaluated to indicate whether or not a
panelist meets the matrix criterion. Each of the example panelists
102 corresponds to a panelist vector defined by the indicators
placed in the location (e.g., the row) of the indicator matrix
corresponding to the panelist and each of the matrix criteria
evaluated. Each panelist vector contains indicator(s) for all of
the matrix criteria evaluated for each of the panelists 102.
[0048] While an example manner of implementing the example data
partitioner 106 of FIG. 1 is illustrated in FIGS. 1 and/or 2, one
or more of the elements, processes and/or devices illustrated in
FIGS. 1 and/or 2 may be combined, divided, re-arranged, omitted,
eliminated and/or implemented in any other way. Further, the
example parameter definer 202, the example criteria definer 204,
the example element and vector positioner 206, the example
indicator matrix generator 110, the example matrix reducer 112, the
example panelist labeler 114, the example panelist sorter 116,
and/or, more generally, the example data partitioner 106 of FIG. 1
may be implemented by hardware, software, firmware and/or any
combination of hardware, software and/or firmware. Thus, for
example, any of the example parameter definer 202, the example
criteria definer 204, the example element and vector positioner 206
the example indicator matrix generator 110, the example matrix
reducer 112, the example panelist labeler 114, the example panelist
sorter 116, and/or, more generally, the example data partitioner
106 could be implemented by one or more analog or digital
circuit(s), logic circuits, programmable processor(s), application
specific integrated circuit(s) (ASIC(s)), programmable logic
device(s) (PLD(s)) and/or field programmable logic device(s)
(FPLD(s)). When reading any of the apparatus or system claims of
this patent to cover a purely software and/or firmware
implementation, at least one of the example parameter definer 202,
the example criteria definer 204, the example element and vector
positioner 206 the example indicator matrix generator 110, the
example matrix reducer 112, the example panelist labeler 114, the
example panelist sorter 116, and/or, more generally, the example
data partitioner 106 is/are hereby expressly defined to include a
tangible computer readable storage device or storage disk such as a
memory, a digital versatile disk (DVD), a compact disk (CD), a
Blu-ray disk, etc. storing the software and/or firmware. Further
still, the example data partitioner 106 of FIG. 1 may include one
or more elements, processes and/or devices in addition to, or
instead of, those illustrated in FIG. 2, and/or may include more
than one of any or all of the illustrated elements, processes and
devices.
[0049] A flowchart representative of example machine readable
instructions for implementing the example data partitioner 106 of
FIGS. 1 and/or 2 is shown in FIGS. 3 and 4. In this example, the
machine readable instructions comprise a program for execution by a
processor such as the processor 1112 shown in the example processor
platform 1100 discussed below in connection with FIG. 11. The
program may be embodied in software stored on a tangible computer
readable storage medium such as a CD-ROM, a floppy disk, a hard
drive, a digital versatile disk (DVD), a Blu-ray disk, or a memory
associated with the processor 1112, but the entire program and/or
parts thereof could alternatively be executed by a device other
than the processor 1112 and/or embodied in firmware or dedicated
hardware. Further, although the example program is described with
reference to the flowchart illustrated in FIGS. 3 and/or 4, many
other methods of implementing the example data partitioner 106 may
alternatively be used. For example, the order of execution of the
blocks may be changed, and/or some of the blocks described may be
changed, eliminated, or combined.
[0050] As mentioned above, the example processes of FIGS. 3 and 4
may be implemented using coded instructions (e.g., computer and/or
machine readable instructions) stored on a tangible computer
readable storage medium such as a hard disk drive, a flash memory,
a read-only memory (ROM), a compact disk (CD), a digital versatile
disk (DVD), a cache, a random-access memory (RAM) and/or any other
storage device or storage disk in which information is stored for
any duration (e.g., for extended time periods, permanently, for
brief instances, for temporarily buffering, and/or for caching of
the information). As used herein, the term tangible computer
readable storage medium is expressly defined to include any type of
computer readable storage device and/or storage disk and to exclude
propagating signals and transmission media. As used herein,
"tangible computer readable storage medium" and "tangible machine
readable storage medium" are used interchangeably. Additionally or
alternatively, the example processes of FIGS. 3 and 4 may be
implemented using coded instructions (e.g., computer and/or machine
readable instructions) stored on a non-transitory computer and/or
machine readable medium such as a hard disk drive, a flash memory,
a read-only memory, a compact disk, a digital versatile disk, a
cache, a random-access memory and/or any other storage device or
storage disk in which information is stored for any duration (e.g.,
for extended time periods, permanently, for brief instances, for
temporarily buffering, and/or for caching of the information). As
used herein, the term non-transitory computer readable medium is
expressly defined to include any type of computer readable storage
device and/or storage disk and to exclude propagating signals and
transmission media. As used herein, when the phrase "at least" is
used as the transition term in a preamble of a claim, it is
open-ended in the same manner as the term "comprising" is open
ended.
[0051] FIG. 3 is a flowchart representative of example machine
readable instructions 300 that may be executed to implement the
example data partitioner 106 of FIGS. 1 and/or 2. The instructions
begin when the example indicator matrix generator 110 of the
example data partitioner 106 receives panelist data from the
example panelist data storage device 104 (block 302). The example
indicator matrix generator 110 generates an indicator matrix based
on the panelist data and the supplied matrix criteria (e.g., matrix
criteria that includes probabilities input by a survey
administrator) (block 304). Example instructions that are executed
to generate an indicator matrix are described below in connection
with the flowchart of FIG. 4. Based on the generated indicator
matrix, a unique set of mutually exclusive partitions is determined
by the example matrix reducer 112 (block 306). The example panelist
labeler 114 assigns a unique label to each of the mutually
exclusive partitions (block 308). The example panelist sorter 116
matches one of the example panelists 102 to corresponding ones of
the example mutually exclusive partitions (block 310). The example
panelist labeler 114 assigns each of the panelists 102 a label
corresponding to the mutually exclusive partition to which the
panelist was matched (block 312). The example panelist sorter 116
sorts panelists according to the assigned label for later use such
as, for example, weighting the partitions (block 314). In some
examples, the full data may be discarded in favor of the summarized
data for the partitions (block 316). In other examples, the
reduction of data may not be performed. The program of FIG. 3 is
then complete.
[0052] FIG. 4 is a flowchart representative of example machine
readable instructions 304 that may be executed to generate an
indicator matrix for received matrix criteria and a set of panelist
data (e.g., data for the example panelists 102). The example
instructions 304 begin with the example parameter definer 202
defining the matrix parameters of the indicator matrix based on the
panelists 102 and the given matrix criteria (block 402). The
example criteria analyzer 204 determines if a first panelist of the
plurality of panelists 102 meets an event of a matrix criterion by
determining if the first panelist fulfills the condition (block
404). The example element and vector positioner 206 places an
indicator in the location of the indicator matrix (e.g., the first
element vector for the matrix criterion) corresponding to the first
panelist and the event of the matrix criterion that indicates
whether or not the first panelist meets the event of the matrix
criterion (block 406). The example criteria analyzer 204 determines
if the first panelist meets a condition of the matrix criterion by
determining if the first panelist fulfills the event and the
condition of the matrix criterion (block 408). The example element
and vector positioner 206 places an indicator in the location of
the indicator matrix (e.g., the second element vector for the
matrix criterion) corresponding to the panelist and the condition
of the matrix criterion that indicates whether or not the first
panelist meets the condition of the matrix criterion (block
410).
[0053] The example indicator matrix generator 110 determines if
there are more matrix criteria to be evaluated for the first
panelist (block 412). If more matrix criteria exist, control
returns to block 404 to evaluate the first panelist based on a next
matrix criterion. If no further matrix criteria exist, the example
indicator matrix generator 110 determines if there are more of the
panelists 102 to be evaluated (block 414). If there are more of the
panelists 102 to be evaluated, control returns to block 404 to
evaluate the next panelist based on the matrix criteria. If no
further panelists 102 are to be evaluated, the indicator matrix is
complete, and control returns to block 306 of the flowchart of FIG.
3 to process the indicator matrix.
[0054] FIG. 5 is an example set of matrix criteria 500 that may be
received by the example indicator matrix generator 110. For
example, the example set of matrix criteria 500 may be received via
input from an operator. As shown in FIG. 5, the example set of
matrix criteria 500 is given in a probability format. For example,
for a matrix criterion 502 (e.g., P(Male|Cell Phone)), the example
criteria analyzer 204 evaluates each data corresponding to each of
the panelists 102 to determine which the panelists 102 has a cell
phone (e.g., the condition), then determines which of the set of
panelists with cell phones are male (e.g., the event). As a further
example, another matrix criterion 504 reads P(Male), which does not
include a condition. In such examples, the condition is understood
to include all panelists in the set of panelists. In examples where
there is only one characteristic (e.g., the event) listed, the
example criteria analyzer 204 records that the panelist is one of
the panelists of the set of panelists 102 being analyzed, then
determines if the panelist has the listed characteristic (e.g., if
the panelist is male). Other sets of the matrix criteria 500 may
include any number of matrix criteria relating to any kind of data
or characteristic of the set of panelists 102 to be evaluated.
[0055] FIG. 6 is an example table 600 including an example set of
panelists 602 and an example indicator matrix 604 that may be
generated by the example indicator matrix generator 110 of FIGS. 1
and/or 2. The example indicator matrix 604 has been determined
using the example set of matrix criteria 500 of FIG. 5 and an
example set of panelists 602. Each panelist of the set of panelists
602 includes an example identification number 606, example gender
data 608 designating the panelist's gender, and example cell phone
data 610 designating whether the panelist has a cell phone. The
element vectors 612-634 correspond to the matrix criteria of FIG. 5
and are populated with indicators 636 and 638 based on whether the
each of the example plurality of panelists 602 meets each of the
set of matrix criteria 500. Panelist vectors, such as the panelist
vector 640 for the first panelist, are created by the row
indicators corresponding to each of the panelists. For example,
element vectors 620 and 622 (e.g., columns N3 and D3) correspond to
the matrix criterion 502 shown in FIG. 5.
[0056] Each matrix criterion of the set of matrix criteria 500
corresponds to two element vectors 612-634 in the example indicator
matrix 604. For example, element vector 620 (e.g., N3) corresponds
to the event of matrix criterion 502 of the set of matrix criteria
500 in FIG. 5 and element vector 622 (e.g., D3) corresponds to the
condition of matrix criterion 502 of the set of matrix criteria 500
in FIG. 5. The indicator in matrix criterion element vector 620
relates to whether the event and condition of matrix criterion 502
(e.g., male and cell phone) are met and the indicator in matrix
criteria element vector 622 relates to whether the condition of
matrix criterion 502 (e.g., cell phone) is met. Thus, because a
panelist 642 (e.g., the panelists corresponding to the
identification number 606 of "10") is female and has a cell phone,
an indicator of "1" 636 is placed in a location of the element
vector 622 (D3) corresponding to panelist 10 because panelist 10
fulfils the condition of the matrix criterion 502 (e.g., having a
cell phone). An indicator of "0" 638 is placed in a location of the
element vector 620 (N3) corresponding to panelist 10 because
panelist 10 does not also fulfill the event of the matrix criterion
502 (e.g., being male). Each panelist of the example set of
panelists 602 is evaluated based on the example set of matrix
criteria 500 and the determined indicators are placed in the
corresponding locations of element vectors 612-634 to complete the
example indicator matrix 604. In other examples, the panelists 602
may correspond to columns and the matrix criteria may correspond to
rows. Any number of panelists 602 may be added to the indicator
matrix and evaluated based on the example set of matrix criteria
500.
[0057] FIG. 7 is an example set of partitioned data 700 that may be
determined using the examples described herein. The example set of
partitioned data 700 in the illustrated example is the minimum set
of mutually exclusive partitions of the indicators in the element
vectors 612-634 of the example indicator matrix 604 of FIG. 6.
Thus, the example partitioned data 700 is the minimum set of
mutually exclusive partitions that can be used to represent the
panelist data in evaluating the example set of matrix criteria 500.
A label column 702 has been added to the set of partitioned data
700 for the labels 704-710. The labels for the example set of
partitioned data 700 are letters (e.g., A, B, C, and D), but in
other examples, other types of labels may be used.
[0058] FIG. 8 is an example set of panelist data 800 that is
matched in accordance with the partitions of FIG. 7. The example
identification number 606, gender data 608, and cell phone data 610
are present in the example set of panelist data in the illustrated
example. In other examples, the example set of panelist data that
is matched in accordance with the partitions of FIG. 7 may only
include the identification number 606. The panelists 602 have been
assigned a label corresponding to the mutually exclusive partition
of the example set of partitioned data 700 of FIG. 7 to which the
panelist was matched.
[0059] FIG. 9 is an example set of panelist data 900 that is sorted
in accordance with the partitions of FIG. 7. The example set of
panelist data 900 illustrated in FIG. 9 has been sorted based on
the panelists 602 that have a particular label 702. In this
example, the label is "B," but other labels may be used in other
examples.
[0060] FIG. 10 is an example set of partition descriptions 1000 in
accordance with the examples described herein. The description
column 1002 includes example partition descriptions 1004-1010,
which are determined based on the characteristics of the panelists
602 corresponding the each label 702. In other examples, the
descriptions may be based on other characteristics included in the
panelist data.
[0061] FIG. 11 is a block diagram of an example processor platform
1100 capable of executing instructions to implement at least a
portion of the methods 300 and 304 of FIGS. 3 and 4. The processor
platform 1100 can be, for example, a server, a personal computer, a
mobile device (e.g., a cell phone, a smart phone, a tablet such as
an iPad.TM.), a personal digital assistant (PDA), an Internet
appliance or any other type of computing device.
[0062] The processor platform 1100 of the illustrated example
includes a processor 1112. The processor 1112 of the illustrated
example is hardware. For example, the processor 1112 can be
implemented by one or more integrated circuits, logic circuits,
microprocessors or controllers from any desired family or
manufacturer. In this example implementation, the processor 1112
includes the indicator matrix generator 106, the matrix reducer
108, the panelist labeler 110 and the panelist sorter 112. The
indicator matrix generator 106 further includes the parameter
definer 202, the criteria analyzer 204, and the element and vector
positioner 206.
[0063] The processor 1112 of the illustrated example includes a
local memory 1113 (e.g., a cache). The processor 1112 of the
illustrated example is in communication with a main memory
including a volatile memory 1114 and a non-volatile memory 1116 via
a bus 1118. The volatile memory 1114 may be implemented by
Synchronous Dynamic Random Access Memory (SDRAM), Dynamic Random
Access Memory (DRAM), RAMBUS Dynamic Random Access Memory (RDRAM)
and/or any other type of random access memory device. The
non-volatile memory 1116 may be implemented by flash memory and/or
any other desired type of memory device. Access to the main memory
1114, 1116 is controlled by a memory controller.
[0064] The processor platform 1100 of the illustrated example also
includes an interface circuit 1120. The interface circuit 1120 may
be implemented by any type of interface standard, such as an
Ethernet interface, a universal serial bus (USB), and/or a PCI
express interface.
[0065] In the illustrated example, one or more input devices 1122
are connected to the interface circuit 1120. The input device(s)
1122 permit(s) a user to enter data and commands into the processor
1112. The input device(s) 1122 can be implemented by, for example,
an audio sensor, a microphone, a camera (still or video), a
keyboard, a button, a mouse, a touchscreen, a track-pad, a
trackball, isopoint and/or a voice recognition system.
[0066] One or more output devices 1124 are also connected to the
interface circuit 1120 of the illustrated example. The output
devices 1124 can be implemented, for example, by display devices
(e.g., a light emitting diode (LED), an organic light emitting
diode (OLED), a liquid crystal display, a cathode ray tube display
(CRT), a touchscreen, a tactile output device, a printer and/or
speakers). The interface circuit 1120 of the illustrated example,
thus, typically includes a graphics driver card, a graphics driver
chip or a graphics driver processor.
[0067] The interface circuit 1120 of the illustrated example also
includes a communication device such as a transmitter, a receiver,
a transceiver, a modem and/or network interface card to facilitate
exchange of data with external machines (e.g., computing devices of
any kind) via a network 1126 (e.g., an Ethernet connection, a
digital subscriber line (DSL), a telephone line, coaxial cable, a
cellular telephone system, etc.).
[0068] The processor platform 1100 of the illustrated example also
includes one or more mass storage devices 1128 for storing software
and/or data. Examples of such mass storage devices 688 include
floppy disk drives, hard drive disks, compact disk drives, Blu-ray
disk drives, RAID systems, and digital versatile disk (DVD) drives.
One particular example of a mass storage device is the panelist
data storage 104.
[0069] Coded instructions 1132 to implement at least a portion of
the methods 300 and 304 of FIGS. 3 and 4 may be stored in the mass
storage device 628, in the volatile memory 1114, in the
non-volatile memory 1116, and/or on a removable tangible computer
readable storage medium such as a CD or DVD.
[0070] Above disclosed methods, apparatus and articles of
manufacture provide the benefit of reducing the amount of data that
must be stored in relation to a set of panelist data, thus reducing
burdens on a processor and/or memory. For example, instead of a
larger amount of data being stored for each panelist (e.g., gender,
whether the panelist has a cell phone, geographic location, etc.),
a label corresponding to a partition (e.g., a partition from the
minimum set of mutually exclusive partitions defined from the data
set) is stored and the full set of data may be discarded thereby
greatly reducing storage space usage. In contrast to previous
methods of determining mutually exclusive partitions for a set of
data that determine all possible mutually exclusive partitions,
some example methods and apparatus disclosed herein determine the
minimum set of mutually exclusive partitions of the data set that
can satisfy the matrix criteria (e.g., provide answers to the
matrix criteria probabilities), which further reduces the amount of
data stored and the burden on the processor. For example, a data
set may contain data related to 10,000 panelists and 4 mutually
exclusive partitions may be defined from the data set. In this
example, the data corresponding to the 10,000 panelists, which may
include 2 different characteristics, can be reduced to 4
partitions, which may include a partition label and a number of
panelists in the partition, thus reducing the amount of information
stored from 20,000 data elements (e.g., 2 characteristics for each
of 10,000 panelists) to 4 data elements. In some examples, a
description may be stored in addition to the label. Simply storing
a label indicating a partition to which the panelist belongs can
expedite future data processing for the set of panelist data
because the full set of data will not have to be re-evaluated for
use. Moreover, the size of the data set to be processed is reduced,
thereby reducing the usage of computer storage and the burden on
the processor and improving the overall operating efficiency of the
corresponding computer.
[0071] Although certain example methods, apparatus and articles of
manufacture have been disclosed herein, the scope of coverage of
this patent is not limited thereto. On the contrary, this patent
covers all methods, apparatus and articles of manufacture fairly
falling within the scope of the claims of this patent.
* * * * *